The cost of treating complications of severe open and closed long bone fractures amounts to more than a billion dollars a year. Many of these fractures are managed with external fixators which permit optimal care of associated soft tissue injuries, but provide mechanical conditions neither ideal for primary nor secondary bone healing. It is the long- term goal of this proposal to decrease healing times and healing failures in complex fractures by optimizing the mechanical properties of external fixators. While it is accepted that fixator rigidity and fracture consolidation are related, clinical and experimental attempts to determine optimal fixator rigidities have led to equivocal results. It is our hypothesis that: 1) fixator rigidities substantially lower than the bending stiffness of the corresponding long bone will result in a more rapid increase in callus stiffness, faster callus proliferation and shorter healing times than the use of stiffer external fixators, and that (2) optimal fixator rigidity in the early part of fracture healing is different from optimal rigidity in the latter part of the healing process. To test these hypotheses and further develop our assessment methods, we have designed four studies: (1) We shall immobilize the tibia in a canine fracture model with 4 fixators of widely varying stiffness. The healing process is assessed weekly by recording changes in callus volume radiographically and changes in callus stiffness with a 6-degrees-of-freedom instrumented external fixator in conjunction with a finite element model (FEA) of the fixator-bone construct. The resulting curves of changing callus dimensions reflect the intensity of callus proliferation and the curves of changing callus stiffness allow us to determine the time relationship of fixator stiffness to callus consolidation. (2) Based on the findings of the initial study, we shall determine the in vivo healing curves in two fixator regimens that exchange low fixator stiffness in the early part of the healing course with high fixator stiffness later and vice versa. (3) With the help of the same technology, we shall determine in vivo healing curves in humans with severe tibial fractures and develop a protocol to optimize fixator stiffness in humans. (4) We shall enhance the performance of the FEA model by incorporating nonlinear behavior and further develop it to gain insight into problems at the pin-bone interface and the mechanics of complex fractures. We believe that the combination of a carefully controlled animal model, the ability to make repeated, sensitive in vivo measurements of callus volume and callus stiffness, and the use of a finite element model that represents the whole fixator-bone construct, will allow us to clarify the long-standing, but clinically important issue of optimal fixator stiffness for the treatment of severe fractures. Later we shall adapt the same technology to explore how other mechanical, metabolic or pharmacological modalities can further accelerate the fracture healing process.